The ultimate embodiment of the first extracorporeal closed-loop BG (BG) control system, or bionic endocrine pancreas, will likely be a portable device consisting of three essential components: a continuous glucose monitor (CGM), a continuous subcutaneous (SC) drug infusion device (e.g. an insulin pump), and a control algorithm. The control algorithm links the other two components by receiving the glucose data stream from the CGM, making a therapeutic decision based on that data stream, and issuing control doses to the drug infusion device. We have designed and extensively tested our closed-loop BG control system in preclinical studies in diabetic swine at Boston University (BU) and in clinical feasibility studies in pediatric and adult subjects with type 1 diabetes in the Clinical Research Center (CRC) at the Massachusetts General Hospital (MGH). In our preclinical and clinical studies, we have tested both insulin-only and bihormonal (insulin and glucagon) configurations of our system. Although our current platform, which runs our control algorithm on a laptop computer, is mobile, it allows our subjects only limited mobility as they are physically tethered to within two feet of an IV pole at all times. To overcome this limitation, we have built the first truly mobile platform for testing a bionic endocrine pancreas in transitional studies in he out-patient setting. The system is built around an iPhone 4S, which runs our control algorithm and coordinates all communication between the CGM, control algorithm, and insulin pump(s). Our new mobile-device platform will allow us to transition from CRC-based feasibility studies to out-patient transitional studies. Furthermore, our new platform is highly flexible and will allow u to test our system in a variety of study cohorts (e.g. new diagnoses, C-peptide positive subjects, pre-adolescent subjects, etc), and with new sensor and infusion technologies, as they become available. As such, our new platform provides us with an enduring research tool that will allow us to test generations of new technologies, study new cohorts, and add progressive improvements in form and function to our bionic pancreas. Our objective is to use our new mobile-device platform to conduct transitional studies in the out-patient setting and to converge upon final device specifications (including algorithm architecture, parametric settings, and user-interface design) for integration into what will become the final embodiment of a commercial product that can be built and then tested in a future pivotal study. We propose to achieve this objective using our new mobile-device platform with the following three specific aims: (1) to conduct an out-patient transitional study testing our device in 32 pediatric subjects (7-21 years old) with type 1 diabetes at the Barton Center's two diabetes camp facilities in Massachusetts, (2) to conduct an out-patient transitional study testing our device in 12 adult subjects (21 years or older) chosen from among the staff at MGH with type 1 diabetes in the setting of their routine work week at MGH, and (3) to conduct an out-patient transitional study testing our device in 18 subjects (7 years or older) who are within three months of diagnosis with type 1 diabetes and who have a positive stimulated C-peptide.